Self-Supervised Learning Based Handwriting Verification

We present SSL-HV: Self-Supervised Learning approaches for the task of Handwriting Verification. This task involves determining whether a given pair of handwritten images originate from the same or different writer distribution. We have compared the performance of four generative and eight contrastive SSL approaches against handcrafted feature extractors and supervised approaches on CEDAR AND dataset. We show that ResNet based Variational Auto-Encoder (VAE) outperforms other generative approaches achieving 76.3% accuracy, while ResNet-18 fine-tuned using Variance-Invariance-Covariance Regularization (VICReg) outperforms other contrastive approaches achieving 78% accuracy. Using a pre-trained VAE and VICReg for the downstream task of writer verification we observed a relative improvement in accuracy of 6.7% and 9% over ResNet-18 supervised baseline with 10% writer labels. Our code is publicly available at: https://github.com/Mihir2/ssl-hv.